About the Sales Charts

Each week, sometime before 8am EST Monday morning and often early on Sunday afternoon, I post a set of comparative rankings that purport to track online sales. At least, that’s what I think my spreadsheet is doing and I spend a fair bit of time doing it.

I don’t have Neilsen BookScan numbers, and likely never will. I rely on sales ‘charts’ (rankings from online sales sites) to try and estimate how manga titles are doing compared to each other. My source data are rankings based on online sales — just a fraction of the total sales, and one that may vary as a percentage of total sales by quite a bit, from publisher to publisher and even from title to title. Another source of errors are the sites themselves: ‘bestseller’ rankings may reflect historical performance in addition to actual sales for any given recent time period (I don’t know how long of a time frame each site considers) and the bestseller formula used will also be different from site to site.

My weekly compilations are determined using 38 different “bestselling manga” lists pulled from 10 different online sales sites over a two week period, and assigning points based on how titles rank. Higher rankings = higher scores.

“Bestselling” is in quotes because I have no idea how a given site chooses to rank their manga; “manga” is in quotes because I use the categories and search results provided by each site. I reserve the right to skip past something that obviously isn’t manga (i.e. Tokyopop’s Hannah Montana) but some stuff (i.e. Thompson’s Manga: The Complete Guide) is worth listing anyway.

Bookstores’ sales sites are given extra weight — not only do we look deeper into the category (a top 300 as opposed to a top 50 or 100), but B&N and Borders are also checked multiple times a week. I consider:

A top 300 for B&N, Borders, Books-a-Million, and Chapters (&Amazon)
A top 100 for Buy.com, DeepDiscount.com, Powell’s, Tower, and Virgin (&Amazon)

Amazon’s Manga Category, their Hourly Bestsellers, B&N, and Borders are all checked three times a week. Other sites are only checked once. Yes, Amazon gets to have it both ways: a top 300 (manga category listings) just like a bookstore, and several top 100 lists (hourly bestseller rankings) like the second-tier sales sites. This is not because I like Amazon–quite the contrary–but rather because they are too big to ignore. And despite your assumptions and any logic that might be applied to their sales reporting, the Amazon Hourly Bestsellers list for manga is not the same as a listing of ‘bestselling’ books in Amazon’s manga category. Check for yourself. There are other complaints I could make about Amazon, but since I do so at least twice a month in the regular posts I’m sure you’ll catch up quickly even without a summary on this page.
For the bestsellers found at each site on a given occasion, I award 1 point for placing last, and an additional point for each rank above that. –for example, for one of the top 300 lists, #300 gets 1 point, #204 gets 97 points, #151 gets 150 points, #150 gets… 151 points, #98 gets [wait for it] 203 points, and #1 gets the whole 300.

The composition of my sources has varied a bit over time, and included one or two other sites that have since been dropped for one reason or another, but that’s the current mix

Once we have all the sales rankings from the varied sites, and have awarded points, the 9000 or so separate listings are tossed into a spreadsheet along with the full, final rankings (around 1500 volumes) from the previous week and through esoteric thaumaturgical processes we get totals for each volume and, eventually, a Top 500 list.

It all starts with the Top 500: Those rankings (and other data from the sources listed below) are then used to compile the derivative charts:
the Top 50 Series
A Publishers’ Scorecard, and
the Manga Midlist 500.

Other ways to look at the numbers (along with anything else I think of) end up in a weekly Sales Commentary post.

##

The secondary charts:

The Top 50 Series chart uses the same scores assigned to books for the Manga Top 500, but with a sprinkling of extra math: A weighted score is determined using the points from the top ranked volume of a given series as a base, and only adding one tenth of the scores for all other books in the series. The Top 50 chart pulls in data not just from the Top 500, but from the full list (1293 volumes this week).

The Publisher’s Scorecard is the most straightforward of the lot: just look at the Top 500 and count: so many for Viz, so many for Tokyopop, etc. Actually, I get the spreadsheet to count them for me, but that’s the gist of it.

The “Midlist 500″ is a re-ranking of manga volumes after excluding all books from the top 5 Series: At the time of this posting (17 April 08) the top 5 series are Naruto, Death Note, Bleach, Fruits Basket, and Fullmetal Alchemist; all together this represents some 117 books of which 110 are clogging up the Top 500. After excluding these volumes I then re-run and re-number the chart with the books that are left.

The Midlist 500 is the reason I set up the spreadsheet and do the rest of the math.

Links to Sources
Online Bookstores:
Barnes & Noble, Books-a-Million, Borders (beta), Chapters of Canada, and Amazon’s Manga sales by Category
Other Sales Charts:
Amazon Hourly Bestsellers, Buy.com, Deepdiscount.com, Fye*, Powell’s, Tower, and Virgin

*dropped May ‘08